A study into the different mechanisms of anti cancer effect of docosahexanoic acid in colon cancer cells

A STUDY INTO THE DIFFERENT MECHANISMS OF CANCER EFFECT OF DOCOSAHEXANOIC ACID IN COLON CANCER CELLS MANAV DEPARTMENT OF COMMUNITY, OCCUPATIOBNAL AND FAMILY MEDICINE NATIONAL UNIVERS

A STUDY INTO THE DIFFERENT MECHANISMS OF CANCER EFFECT OF DOCOSAHEXANOIC ACID

IN COLON CANCER CELLS

MANAV

DEPARTMENT OF COMMUNITY, OCCUPATIOBNAL

AND FAMILY MEDICINE NATIONAL UNIVERSITY OF SINGAPORE

2004

A STUDY INTO THE DIFFERENT MECHANISMS OF CANCER EFFECT OF DOCOSAHEXANOIC ACID

IN COLON CANCER CELLS

MANAV (B MED, GSMU, INDIA)

A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF SCIENCE

DEPARTMENT OF COMMUNITY, OCCUPATIOBNAL

AND FAMILY MEDICINE NATIONAL UNIVERSITY OF SINGAPORE

2004

ACKNOWLEDGEMENTS

I would like to express my respect and gratitude to Professor Ong Choon Nam and Professor Lee Hin Peng As my supervisors, they ensured that I remained focused on achieving my goal Their observations and guidance helped me to establish the overall direction of the research and to move forward with investigation in depth What I learned from them, especially their approach to a scientific question, is an invaluable lesson for

me not only in the academic perspective but also in my personal life

I would also like to express my sincere thanks to:

Prof David Koh, Head of the Department, for his support during the course of the study;

Dr Shen Han Ming, for his advice and stimulating discussions;

Mr Ong Her Yam, Madam Lee Bee Lan, Madam Jin Su, Madam New Ai Li and Mr Ong Yeong Bing and Madam Zhao Min, for their guidance and help in the laboratory work.;

My seniors and colleagues Zhang Siyuan, Won Yen Kim, Shi Ranxin, Huang Qing and

Lu Guodong ; they were responsible for the fine-tuning of my techniques and ideas; National University of Singapore, for providing me a research scholarship for this study

Finally, I thank my family for their never-ending love and support I dedicate this thesis

to them

TABLE OF CONTENTS

Acknowledgements i

Table of contents ii

Abbreviations vii

List of publications ix

Summary x CHAPTER I: INTRODUCTION

1.1 Polyunsaturated Fatty Acids (PUFAs) 2

1.1.1 Structure of PUFAs 2

1.1.2 Synthesis of n-3 and n-6 PUFAs 2

1.1.3 Dietary sources 4

1.1.4 Intestinal absorption, metabolism 4

1.1.5 Functions of PUFAs 5

1.1.6 Competition between n-3 and n-6 PUFAs 6

1.2 PUFAs and colorectal cancer 6 1.2.1 Epidemiological Studies 6

1.2.2 Animal studies 8

1.2.3 Intervention studies in humans 10 1.2.4 In vitro studies 10

1.3 Apoptosis 11

1.3.1 Apoptosis - A brief introduction 11

1.4.1 Reactive Oxygen Species (ROS) – Definition and source 14

1.4.2 Role of oxidative stress in carcinogenesis and apoptosis 15

1.5 Mitogen Activated Protein Kinases (MAPKs) 18

1.5.1 MAPK signaling pathways- Introduction 18

1.6 Peroxisome Proliferator-Activated Receptors (PPARs)

2.2 Cell culture and fatty acid supplementation 27 2.3 Determination of cell viability - MTT assay 27

2.4 Evaluation of cell morphological alterations 28

2.5 Determination of sub-G1 population 28 2.6 Measurement of cellular caspase-3 activity 29

3.2.1.1 Effect of different fatty acids on cell viability of HT-29 cells 35 3.2.1.2 Effect of DHA on cell viability of different colon cancer cell lines 35 3.2.1.3 Effect of metabolic pathway inhibitors on DHA-induced cell death 35 3.2.2 DHA induced cell death is apoptotic 36 3.2.3 DHA induced apoptosis is caspase mediated 37 3.2.3.1 Determination of caspase-3 activation 37 3.2.3.2 Effect of caspase-3 inhibitor on DHA-induced cell death 37

CHAPTER IV: EFFECT OF DHA ON ROS GENERATION AND MAP KINASE

4.1 Introduction 50

4.2.1 Effect of DHA on MAPK signaling pathways 50

4.2.1.2 Effect of MAPK inhibitors on DHA-induced cell death 51 4.2.1.3 Effect of JNK mutants on DHA-induced cell death 52

4.2.2.2 Effect of antioxidants and H202 on DHA-induced MAPK activation 53 4.2.2.3 Effect of antioxidants on DHA-induced cell death 54

CHAPTER V: DHA INHIBITS TNF-α INDUCED COX-2 EXPRESSION AND

NF-ΚB TRANSCRIPTION THROUGH THE PPARγ PATHWAY

5.2.1 DHA inhibits TNF-α induced COX-2 and NF-κB activation 67 5.2.2 DHA induces transcriptional activition of PPARγ 68 5.2.3 DHA regulates COX-2 expression and NF-κB activation through

5.2.4 DHA induces apoptosis in colon cancer cells by a PPARγ-independent

CHAPTER VI: DISCUSSION AND CONCLUSIONS

6.3 Comparison of results from in vitro and in vivo studies 87

ABBREVIATIONS

PPAR Peroxisome Proliferator-Activated Receptor

PPRE Peroxisome Proliferator-activated receptor response

element

LIST OF PUBLICATIONS

• Manav, Su, J., Hughes, K., Lee, H.P., and Ong, C.N (2004) n-3 Fatty Acids and Selenium as Coronary Heart Disease Risk Modifying Factors in Asian Indian and

Chinese Males Nutrition In press

• Manav, HM Shen and CN Ong Involvement of c-Jun N-terminal kinase

activation in Docosahexanoic acid-induced apoptosis in colon cancer cell In preparation

• Manav, HM Shen and CN Ong Docosahexanoic acid inhibits TNF-α induced

COX-2 expression in colon cancer cells: Role of PPARγ In preparation

SUMMARY

Colon cancer is one of the leading causes of cancer mortality in the developed world Evidence from epidemiological and animal studies indicates that the n-3 polyunsaturated fatty acids (PUFAs) may protect against colon cancer However the mechanisms of the anti-cancer effects of n-3 PUFAs have not been fully elucidated Docosahexanoic acid (DHA) (22:6, n-3), abundantly found in fish oil, is one of the principal n-3 PUFAs The aims of this study were (i) to investigate the effects of the n-3 PUFAs, and DHA in particular, on cell growth and apoptosis in colon cancer cell lines (ii) to explore the signaling pathways responsible for the chemotherapeutic and chemopreventive actions of DHA

The growth inhibitory effects of DHA were demonstrated by the MTT assay DHA was more potent in inhibiting growth than the n-6 PUFAs, monounsaturated and saturated fatty acids The cytotoxic effect of DHA was due to apoptosis, as demonstrated

by nuclear condensation, sub-G1 assay and PARP cleavage DHA also activated 3; and the caspase inhibitor z-VAD-fmk was able to block the DHA-induced apoptosis

caspase-The effect of DHA on the mitogen-activated signaling kinases (MAPKs) was also investigated DHA activated JNK, ERK and p38 MAPKs The activation of JNK played a key role in the apoptotic action of DHA, as inhibition of JNK either by the specific JNK inhibitor SP600125 or by overexpression of the dominant-negative mutants of JNK significantly repressed both the JNK activity and apoptosis induced by DHA It was further shown that DHA induces oxidative stress but the JNK activity and apoptosis were independent of the reactive oxygen species generation

DHA also induced the transcriptional activity of the Peroxisome activated receptor γ (PPARγ) Activation of ERK interfered with ability of DHA to activate PPARγ DHA also inhibited TNF-α – induced COX-2 expression and NF-κB activation This inhibition was dependent on PPARγ as repressing the PPARγ expression

Proliferator-by transfection of antisense PPARγ oligonucleotides, decreased the inhibitory effect of DHA on COX-2 expression However, there were no significant changes in cell death when PPARγ was inhibited, indicating the possibility that DHA induced apoptosis through PPARγ-independent pathway

In conclusion, this study provides an insight into the pathways involved in the role

of DHA in colon cancer This study for the first time shows (i) activation of JNK plays a key role in DHA-induced apoptosis (ii) DHA inhibits COX-2 by the activation of PPARγ Thus, different mechanisms seem to be involved in the anti-carcinogenic effect of DHA

in colon cancer

CHAPTER I

INTRODUCTION

1.1 Polyunsaturated Fatty Acids

Polyunsaturated fatty acids (PUFAs) are ubiquitious biological molecules that function as metabolic fuels, as covalent regulators of signaling molecules, and as essential components of cellular membranes However, alongside these essential functions, fatty acids have also been identified as having an effect, both positive as well

as negative, on health and disease progression

1.1.1 Structure

Fats form a large group, composed of different types of fatty acids and glycerol Fatty acids are hydrocarbon chains with a carboxyl group at one end Fatty acids usually contain an even number of carbon atoms and on the basis of their degree of saturation, can be classified into saturated fatty acids (SFA), monounsaturated fatty acids (MUFA) (with one double bond), and polyunsaturated fatty acids, PUFAs (with multiple double bonds) According to conventional nomenclature of fatty acid molecules, PUFAs are classified into different groups on the basis of the position of the first double bond from the methyl terminus of the hydrocarbon chain of the molecule Thus n-3 and n-6 PUFAs are so named as they have their first double bond at the 3rd and 6th carbon respectively (Nettleton, 1995)

1.1.2 Synthesis of n-3 and n-6 PUFAs

Most of the n-3 and n-6 PUFAs are metabolized from their precursors, linoleic acid (LA; 18:2n-6) and alpha-linolenic acid (ALA; 18:3n-3) respectively Both LA and ALA are metabolized to longer chain PUFAs, largely in the liver, by a series of

Fig 1.1 Structure and classification of fatty acids

elongation and desaturation reactions to yield longer, more unsaturated fatty acids LA is converted to AA (20:4n-6) while ALA is converted to eicosapentanic acid (EPA;20:5n-3) and docosahexanoic acid (DHA; 22:6n-3) Mammalian organisms, unlike plants do not possess the ∆12 and ∆15 desaturase enzymes required for LA and ALA synthesis Therefore these two PUFAs are considered as essential and must be supplemented in the diet in order to maintain adequate body pools (Simopoulos, 1999)

1.1.3 Dietary sources of PUFAs

ALA is the major fatty acid in chloroplast lipids, and hence is the major n-3 PUFA from plant derived sources Soybean, canola, perilla, linseed, and rapeseed oils are rich sources of ALA, apart from green leafy plants DHA and EPA are found exclusively

in aquatic animals as they are synthesized by phytoplankton and algae, which is then consumed by fish, mollusks and crustaceans, and thereby concentrated in the aquatic food chain (Nettleton, 1995) The relative amounts of DHA and EPA contained in fish oils vary considerably between species, with deep sea fish containing more of these n-3 PUFAs compared to the freshwater ones (Childs et al., 1990) Among the n-6 PUFAs, LA can be found in high proportions in many vegetable seeds and oils (safflower, soybean, coconut, corn and sunflower) Dietary LA is considered to be the major source of tissue

AA, although lean meats and meat fat are the direct sources in the human diet (Li et al., 1998)

1.1.4 Intestinal absorption and metabolism

Absorption of dietary n-3 PUFAs depends on the form in which they are ingested Fish and fish oil products are mostly ingested as triglycerides In mammals, the relative absorption of different forms of PUFAs varies as Free PUFA > triglyceride > ethyl ester (Nelson and Ackman, 1988) Ingestion of n-3 PUFAs leads to their distribution to virtually every cell in the body The non-esterified fatty acids enter the cells via fatty acid transporters and are rapidly converted to fatty acyl-CoA thioesters (FA-CoA) by acyl-CoA synthetases A major fraction of these lipids is bound to specific proteins, i.e fatty acid binding protein and FA-CoA-binding protein FA-CoAs are substrates for neutral

lipid (triglycerides, cholesterol esters) and polar lipid (phospholipids, sphingolipids and plasmalogens) synthesis (Jump, 2002) On exogenous supplementation, the phospholipid pool comprising of phosphatidylcholine and phosphatidylethanolamine, is the major site

of PUFA incorporation in both cultured normal and transformed cells (de Bravo et al., 1991)

1.1.5 Function of PUFAs

Dietary PUFAs have effects on diverse physiological processes impacting normal health and chronic disease, such as the regulation of plasma lipid levels (Harris, 1997; Mori et al., 2000), cardiovascular (Nilsen and Harris, 2004) and immune function (Hwang, 2000) As structural phospholipids of cell membranes, they modulate membrane fluidity, cellular signaling and cellular interaction Moreover they play an important role

in the regulation of the immune system by acting as precursors for the synthesis of eicosanoids Arachidonic acid (AA) or EPA are mobilized from the cell membrane by the action of the phospholipase enzymes especially phospholipase A2 (PLA2) and C (PLC), and subsequently metabolized by cyclooxygenase (COX) and lipoxygenase (LOX) into prostaglandins (PGs), thromboxanes (TXs) and leukotrienes (LTs) As membrane phospholipids normally contain much higher levels of arachidonic acid (AA) than of the other 20-carbon PUFAs (Yaqoob et al., 2000), AA is the most common eicosanoid precursor and gives rise to 2-series PGs and TXs and 4-series LTs In contrast, EPA gives rise to 3-series PGs and TXs and 5-series LTs, the difference being the presence of the double bond in the structure (Johnson et al., 1983) n-3 PUFAs also play a role in the regulation of gene expression (Jump and Clarke, 1999; Duplus et al., 2000) However

cell-specific lipid metabolism as well as the regulation of fatty acid regulated transcription factors are likely to play an important role in determining how cells respond

to changes in PUFA composition

1.1.6 Competition between n-3 and n-6 PUFAs

There is competition between n-3 and n-6 PUFAs for their metabolic conversion via the desaturase and elongase enzymes, which are common to both pathways However these enzymes have a greater affinity for n-3 PUFAs such that when dietary n-3 intake is high, they are preferentially metabolized (Jump, 2002) This leads to competitive inhibition of n-6 PUFA metabolism, where linoleic acid desaturation and AA concentration are significantly decreased after EPA or DHA supplementation, and resulting in decreased generation of AA-derived eicosanoids (Yaqoob et al., 2000; Caughey et al., 1996; Thies et al., 2001)

1.2 PUFAs in colon cancer

1.2.1 Epidemiological studies

Fat has been the focus of dietary studies on colorectal cancer more than any other component The initial insight into the relationship between dietary fat and cancer came partly from epidemiological studies With a few exceptions (Macquart-Moulin et al., 1986; Tuyns et al., 1987), early case control studies showed a positive association between the risk of colorectal cancer and the intake of fat and meat (Potter and McMichael, 1986; Benito et al., 1990; Graham et al., 1988; Lee et al., 1989; La Vecchia

et al., 1988) Studies examining the effect of the degree of saturation of fats highlighted

saturated fat consumption, as an agent responsible for colorectal cancer (Burnstein, 1993; Woutersen et al., 1999) On the other hand, a decrease in risk of colon cancer was reported with an increase in the degree of unsaturation of fats (Lee et al., 1989; Macquart-Moulin et al., 1987; Benito et al., 1991)

A reportedly lower incidence of thrombotic and immunologically mediated diseases in Greenland Eskimos when compared with mainland Danish population, aroused interest in the potential beneficial effects of marine lipiods (Bjerregaard and Dyerberg, 1988) Lanier et al, (1976) in a 5-year survey from 1969-73 observed an increase in cancer incidence of lung, colon and rectum among the Alaska Eskimos and Aleuts A further survey of cancer incidence for the years 1989-1993 found that there were significant increases in the rates for cancers of prostate and colon in men (Lanier et

al, 1996) The diet of Eskimos of Alaska is high in fat but this comes largely from marine animals and fish (Nobmann et al, 1992) Urbanization of the native population and the decreasing trend of fish intake were suggested as contributing factors to the increase in cancer rates, implying that fish and fish oils may have a protective risk modifying effect

in colorectal cancer In an analysis involving 24 European countries, an inverse correlation was found in males between colorectal cancer mortality and current fish intake There was evidence of a protective effect of a high fish intake relative to that of dietary sources of n-6 PUFAs (Caygill and Hill, 1995) In a follow-up study, mortality from colon cancer correlated with the consumption of animal fat and an inverse correlation was observed with fish and fish oil consumption when expressed as a proportion of total fat (Caygill et al., 1996) Fish and fish oil being rich sources of n-3 PUFAs, the findings of this analysis implied that lower levels of n-3 PUFAs and higher

levels of n-6 fatty acids in the body may be a predisposing factor in the causation of colon cancer

Prospective cohort studies also have shown positive association between fat, meat and colorectal cancer Willett et al, (1990) reported an almost 2-fold higher risk of colorectal cancer among women in the highest quintile of red meat consumption in the Nurses’ study The Health Professionals Follow-Up Study, a cohort study of men, also demonstrated a direct association between red meat consumption and risk of colon cancer, but no association was observed with other sources of fat (Giovannucci et al., 1994) The protective effects of fish consumption are seen only in areas where fish consumption is high Fishermen on the west coast of South Africa, having a significantly higher (110 versus 30g/day) had six times lower colorectal cancer incidence than in white urban dwellers (Schloss et al., 1997) In a study in Norway, intake of fish in general had

no protective effect against colorectal cancer, but the relative risk of people who ate five

or more fish meals per week was lower than that of people who ate fish less frequently (Gaard et al., 1996) Taken together, these results indicate that various types of fat may have opposite effects on the risk for colorectal cancer, with meat rich in n-6 PUFAs promoting cancer and fish rich in n-3 PUFAs being protective

1.2.2 Animal studies

The second line of evidence in understanding the influence of dietary fat on colorectal cancer development came from rodent models Laboratory animal studies provided the evidence that not only the amount of fat but also the types of fat are

important factors in colon cancer development Experiments with diets containing

saturated fatty acids, n-3 PUFA and n-6 PUFAs clearly show that n-3 PUFA rich diets inhibit carcinogenesis (Nelson and Ackman, 1988; Reddy and Maruyama, 1986a; Reddy and Sugie, 1988) whereas saturated fats and n-6 PUFA rich diets enhance tumor production (Reddy and Maeura, 1984; Reddy and Maruyama, 1986b) In addition, the stage of carcinogenesis at which the effect of dietary fat is exerted appears to depend on the fatty acid composition, with the n-3 PUFAs being protective in both the initiation and promotion phase of carcinogenesis (Reddy and Maruyama, 1986b; Reddy et al., 1991) The effects of different n-3 and n-6 fatty acid ratios in experimental colon carcinogenesis were studied by Deschner et al, (1990) The highest ratio of n-3:n-6 PUFAs inhibited epithelial cell proliferation and induced S-phase arrest in the colonic cells, whereas the lowest ratio of n-3:n-6 PUFAs produced the highest tumor incidence in azoxymethane treated rats

Germ line mutations of the murine Apc gene provide a model for human familial

adenomatous polyposis The inhibitory effect of DHA and DHA-enriched fish oil were

also demonstrated in mouse models with a mutation at the Apc gene at codon 716 and

codon 850 respectively (Oshima et al., 1995; Paulsen et al., 1997) The administration of DHA has also been shown to inhibit colon cancer cell metastasis with a reduction in the matrix metalloproteinase activity (Suzuki et al., 1997; Iigo et al., 1997) These studies tend to confirm the epidemiological evidence that n-3 PUFAs are protective, whereas n-6 PUFAs promote cancer formation

1.2.3 Intervention studies in humans

Studies describing direct intervention with n-3 PUFAs in human subjects are not many The relative long periods of fatty acid supplementation makes the dietary intervention studies difficult to conduct Subjects, at high risk for colorectal cancer, receiving fish oil supplementation showed changes in proliferation pattern of the rectal mucosa similar to that observed in the low risk population (Anti et al., 1992; Bartoli et al., 1993) In another study, patients with stage 1 or 2 colon carcinoma or adenomatous polyps did not develop additional polyps after 12 months of n-3 PUFA supplementation (Huang et al., 1996) In contrast, Akedo et al, (1998) reported that Familial Adenomatous Polyposis (FAP) patients supplemented with DHA-enriched fish oil capsules, still progressed to malignant lesions after 12 months Though clinical intervention studies are

more relevant to the human in vivo situation, the existing data are not substantial More

studies are needed to confirm the efficacy of clinical application of DHA in colon cancer

1.2.4 In vitro studies

Compared to the numerous epidemiological and in vivo studies that have been

conducted on colon cancer to determine the role of fatty acids, few studies have tried to look into the effect of PUFAs on colon cancer cell lines.(Tsai et al., 1998) showed that n-

3 PUFAs, DHA and EPA inhibited the proliferation of sigmoid colon cancer transformants while having little effect on normal cells In HCT-116 cells transfected with inducible COX-2, both DHA and EPA inhibited cell proliferation, compared with linoleic acid (Boudreau et al., 2001) In CaCo-2 cells, irrespective of the n-3/n-6 status, the longer chain PUFAs - DHA, EPA and AA were more potent in inhibiting growth than

α-linolenic acid and linoleic acid (Dommels et al., 2003; Nano et al., 2003) In HCA-7 cell line, DHA acts synergistically with the selective COX-2 inhibitor, celecoxib, to inhibit cyclooxygenase-2 and cell proliferation (Swamy et al., 2004) In HT-29 cells DHA has been shown to inhibit growth and induce cell cycle arrest (Chen and Istfan, 2001), but the molecular mechanism of these actions still remain to be elucidated

Apart from its effect on cellular proliferation, DHA also modulates other functions in colon cells In young adult mouse colonic cells DHA, compared with

linoleic acid inhibited Ras localization to the plasma membrane and GTP binding (Collett

et al., 2001) Both n-3 and n-6 PUFAs increased the gap junctional intercellular communication during spontaneous differentiation of Caco-2 (Dommels et al., 2002) DHA increases the metabolism of all-trans-retinoic acid and CYP26 gene expression in intestinal cells (Lampen et al., 2001) and selectively activates RXRalpha relative to n-6 PUFA in colonocytes (Fan et al., 2003) cDNA microarray analysis showed that in CaCo-

2 colon cancer cells, DHA down-regulated the inducible nitric oxide synthase, the prostaglandin family of genes, as well as cyclooxygenase-2 expression and several cell cycle-related genes, whereas it up-regulated genes associated with apoptosis (Narayanan

et al., 2003)

1.3 Apoptosis

1.3.1 Apoptosis – A Brief Introduction

Apoptosis or programmed cell death is a critical component of both normal development and disease (Hengartner, 2000) First described by (Kerr et al., 1972), this distinct type of cell death is characterized by cytoplasm swelling, blebbing of the plasma membrane, chromatin condensation maintenance of organelle integrity, and condensation

and fragmentation of DNA, followed by orderly removal through phagocytosis The importance of apoptotic process can be assessed from the fact that the apoptotic machinery has been highly conserved throughout evolution, with many similarities between phylogenetically divergent groups including invertebrates and humans (Wyllie

et al., 1999) Defects in the apoptotic process can result in many pathological conditions including cancer, Alzheimer’s disease, stroke and Acquired Immuno-deficiency Syndrome (Webb et al., 1997)

One of the main executioners of the apoptotic pathway are the caspases Caspases are cysteine-specific proteases that are expressed as inactive precursors Caspases are activated early in the apoptotic cascade either by (i) processing by an upstream caspase (ii) ligand binding to the death receptors (iii) or association with a regulatory subunit like Apaf-1 (Hengartner, 2000) The initial activation of caspases is amplified by the caspase cascade which also integrates the pro-apoptotic signals Proteolytic cleavage of cellular substrates by caspases largely determines the features of apoptosis The caspase substrates range from the single polypeptide chain enzymes, like polyADP-ribose polymerase, to complex macromolecular structures like the lamin network (Creagh et al., 2003)

Mitochondria not only serve as the major energy source in the living cells, but they can also trigger or amplify the signals that lead to apoptosis (Green and Reed, 1998) Alteration in mitochondrial membrane potential and permeabilisation of the mitochondrial membrane often precede caspase activation and other manifestations of apoptosis (Zamzami and Kroemer, 2001) Induction of permeability transition in the inner mitochondrial membrane may be accompanied by the release of cytochrome c,

Smac/Diablo and AIF, leading to the uncoupling of oxidative phosphorylation and accelerated ATP hydrolysis by mitochondrial ATPase (Lemasters et al., 2002) The mitochondrial pathway is used extensively in response to extracellular signals and internal insults such as DNA damage These diverse response pathways converge on the mitochondria, often through the activation of a pro-apoptotic member of the Bcl-2 family The death-receptor pathway and mitochondrial pathways converge at the level of caspase-3 activation, which initiates the proteolytic cascade that culminates in apoptosis (Green and Reed, 1998)

1.3.2 PUFAs and apoptosis

Both deficiency and excess of PUFAs can influence the extent of apoptosis The reported effects of the n-3 PUFAs, DHA and EPA, on apoptosis have ranged from inhibition to stimulation Apoptotic effects of DHA have been reported in many cancer cell lines such as vascular smooth muscle cells (Shiina et al., 1993), Morris hepatic carcinoma 3924A (Calviello et al., 1998), mammary carcinoma (Minami and Noguchi, 1996), and colon cancer cells (Tsai et al., 1998) In contrast, DHA has also been reported

to exert its protective effect by inhibition of apoptosis DHA protects neuro2A cells from serum deprivation induced apoptosis (Kim et al., 2000) and prevents oxidative stress induced apoptosis in retinal cells (Rotstein et al., 2003) In leukemic cells too, DHA inhibits sphingosine and TNF/ cyclohexamide-induced apoptosis (Yano et al., 2000; Kishida et al., 1998)

Among the n-3 PUFAs, DHA and EPA seem to have different effects on apoptosis Administration of low doses of DHA to ACI/T rats transplanted with Morris

hepatocarcinoma 3924A resulted in doubling of the number of cells undergoing apoptosis, whereas administration of EPA did not affect apoptosis but increased the cellular differention (Calviello et al., 1998) For n-6 PUFAs, arachidonic acid and its metabolites, and conjugated linoleic acid are also known to induce apoptosis (Bergamo et al., 2004; Hawkins et al., 1998) Together these studies indicate that PUFAs induce apoptosis in a variety of cell types and the potency to induce apoptosis varies among the different PUFAs

1.4 Oxidative Stress

1.4.1 Reactive Oxygen Species- Definition and Source

Reactive Oxygen Species (ROS) are constantly generated in all organisms as a result of aerobic metabolism (Wiseman and Halliwell, 1996) A free radical is any species capable of independent existence that contains one or more unpaired electrons (Halliwell, 1987) A compound becomes a free radical by either gaining an additional electron or by losing one Oxygen-centred free radicals are the most common because of the prevalence of oxygen in the biological system and their high reactivity towards various cellular and molecular targets (Halliwell, 1999; Marnett, 2000) ROS includes the hydroxyl radicals (.OH), superoxide anions (O2.-), singlet oxygen (1O2) and hydrogen peroxide (H2O2) Among these, .OH, is the most reactive one responsible for most of the oxidative damage caused by ROS such as lipid peroxidation and oxidative damage (Halliwell, 1999)

In aerobic cells, ROS are generated by several pathways Mitochondria is believed to be the major site of ROS production due to the presence of the electron

transport chain (Cai and Jones, 1999) During the production of ATP, in the electron transport reaction O2 accepts electrons and H+, and gets reduced to water In the process,

a small fraction of electrons are leaked to and transferred to O2, thus resulting in the formation of O2.- (Kamata and Hirata, 1999) In the mitochondria, ROS formation is significantly affected by alterations of mitochondrial lipid composition and by lipoperoxidation products (Mates and Sanchez-Jimenez, 2000) Another source of ROS production is the endoplasmic reticulum where ROS is generated either by donation of electrons from NADPH cytochrome p450 reductase or by the microsomal enzymes (hypoxanthine/xanthine oxidase, NADPH oxidase, lipoxygenase, cyclooxygenase) (Kamata and Hirata, 1999)

1.4.2 Role of oxidative stress in carcinogenesis and apoptosis

Cancer and apoptosis are opposed phenomena but ROS have been widely reported to play a key role in both Low levels of ROS regulate cellular signaling and play an important role in normal cellular proliferation (Kamata and Hirata, 1999) Interestingly, tumor cells are known to contain elevated levels of ROS (Burdon, 1995) Persistent oxidative stress results in accumulation of oxidative damage to several critical biomolecules This eventually results in several biological effects ranging from alterations in signal transduction, and activation of transcription factors NF-kB and AP-1 (Gupta et al., 1999) Oxidative stress can also induce DNA damage, which leads to genomic instability and may thus contribute to cancer progression (Jackson and Loeb, 2001) Compelling evidence for the carcinogenic potential of ROS is the finding that ROS is produced in cells stimulated with growth factors such as EGF and PDGF (Bae et

catalytic subunit of NADPH oxidase) induces superoxide generation This enzyme has been shown to transform NIH3T3 cells and also produce aggressive tumors in athymic

mice (Suh et al., 1999), illustrating the critical role of ROS in vitro and in vivo Thus

ROS are thought to play multiple roles in tumor initiation, progression and maintenance

In contrast to their role in promoting cell growth under non-stress conditions, ROS appears to activate and modulate apoptosis when cells are under stress ROS levels are increased in cells exposed to various stress agents, including anticancer drugs (Jabs, 1999), and they promote apoptosis by stimulating various pro-apoptotic signaling molecules, such as ASK1, JNK and p38 (Benhar et al., 2001; Davis, 2000) ROS also plays a pivotal role in p53 induced apoptosis (Polyak et al., 1997) In addition ROS can act directly on the apoptotic machinery, by accelerating mitochondrial depolarization and dysfunction during the effector phase of apoptosis (Jabs, 1999) Thus ROS signaling serves as a bifunctional regulator of cell proliferation

1.4.3 PUFAs and oxidative stress

Recent independent observations in diverse models show that PUFAs may act as pro-oxidants and suggest an important role for oxidative mechanisms in their effects in biological systems

1.4.3.1 PUFAs and lipid peroxidation

Lipid peroxidation is thought to be one of the major mechanisms of PUFA action

It is hypothesized that the presence of double bonds makes the PUFAs extremely sensitive to oxidation Membrane phospholipids are pro-oxidised at the allylic position

between double bonds of unsaturated free fatty acids, resulting in the generation of lipid peroxides (Girotti, 1998) Lipid peroxidation begins with the removal of a hydrogen atom from a PUFA double bond, forming a lipid radical This lipid radical is extremely reactive and reacts with molecular oxygen to form a lipid peroxy radical which, in turn, attacks another PUFA and propagates further reactions (Rice-Evans and Burdon, 1993) Thus a single initiation event results in a chain reaction The oxidative modification of fatty acids results in changes in membrane fluidity, inactivation of membrane bound receptors and enzymes, which may lead to impairment of membrane function and cell death In addition, the lipid peroxides can also be mutagenic and genotoxic by causing damage to DNA and other macromolecules (Leuratti et al., 2002; Marnett, 2000)

1.4.3.2 PUFAs and Reactive Oxygen Species

Another means of PUFA induced oxidative stress is through the production of oxidizing agents such as hydrogen peroxide (Madhavi et al., 1994; Sagar et al., 1992) PUFA can be metabolized by fatty acyl-CoA oxidase, an enzyme in the β-oxidation pathway In mitochondria, these fatty acid metabolites supply reducing equivalents to the redox reactions at complexes I and II Cells overexpressing acyl-CoA oxidase showed that PUFA treatment produced hydrogen peroxide which was accompanied by a decrease

in cellular proliferation (Okamoto et al., 1997) Apart from peroxides, superoxide anion and hydroxyl radical, have also in some instances been associated with the action of PUFAs (Anel et al., 1992; Kumar and Das, 1995) On the other hand, fatty acids are mild uncouplers that act through a protonophoric effect or through activation of uncoupling proteins (UCP) or/ and of the adenine nucleotide translocase (Wojtczak and Schonfeld,

1993; Hermesh et al., 1998) and may reduce mitochondrial hydrogen peroxide generation (Korshunov et al., 1998) Fatty acids can also contribute to the cellular antioxidant defences against mitochondrial oxidative stress by activating glutathione peroxidase through an EGFR-dependent function (Duval et al., 2002) Thus, in the cellular system fatty acids may deal with oxidative stress in different ways

1.5 Mitogen Activated Protein Kinases

1.5.1 Mitogen Activated Protein Kinase (MAPK) signaling pathways: Introduction

Mitogen Activated Protein Kinases (MAPKs) constitute an important group of signaling mediators Signal transduction via MAPKs plays a key role in a variety of cellular responses including growth factor-proliferation, differentiation and cell death

(Cano and Mahadevan, 1995; Robinson and Cobb, 1997) Three major subfamilies of

structurally related MAPKs have been identified in mammalian cells These are the extracellular signal regulated kinases (ERKs), the c-Jun N-terminal kinases/ stress activated protein kinases (JNK/SAPKs), and the p38 kinases MAPK subfamilies phosphorylate their substrates at serine and threonine residues located adjacent to a proline residue, and their substrates include diverse cellular molecules, including protein kinases and transcription factors (Robinson and Cobb, 1997; Cano and Mahadevan, 1995) Members of all three MAPK subfamilies are activated by upstream kinases (MAPK Kinases, or MKKs) by dual phosphorylation of threonine and tyrosine residues

ERKs are characteristically activated by mitogenic stimuli such as growth factors, and are associated with cell proliferation, differentiation and protection from apoptosis

In contrast, the JNK and p38 pathways are strongly activated in response to stress stimuli

such as UV irradiation, hydrogen peroxide, DNA damage, heat and osmotic shock, and are involved in growth arrest and the induction of apoptosis (Kyriakis et al., 1994; Raingeaud et al., 1995; Minden and Karin, 1997) The specificity of activating stimuli for these three subfamilies of MAPKs is not absolute, and much of the overlapping in their roles could be due to different roles of each pathway in various cell types responding to different stimuli

1.5.2 MAPKs and ROS

Recent emerging evidence seems to indicate that redox regulation is important in MAPK activation under stress (Adler et al., 1999) It has been demonstrated that ROS functions as an intermediate in MAPK activation in response to stress agents such as ceramide and anti-cancer drugs (Mansat-de Mas et al., 1999; Shiah et al., 1999) ROS-dependant MAPK activation has also been implicated in regulating the behaviour of tumor cells Activation of MAPKs in transformed murine cells was shown to be independent of the oncogenic transformation but dependant on ROS production (Benhar

et al., 2001) In human tumor cells also, higher ROS levels and subsequent MAPK activity increase the sensitivity to anti-cancer drugs (Benhar et al., 2001) Recently, Sakon et al., (2003) and Chen et al., (2003) demonstrated that increased accumulation of ROS prolonged the activation of JNK when NF-κB is inhibited This may be due to either the decreased expression of the antioxidant enzymes or the increased expression of the p450 genes In another case, overexpression of manganese superoxide dismutase (MnSOD), reduced the oxidative stress and suppressed the JNK/AP-1 pathway in a multistage skin carcinogenesis model (Zhao et al., 2001)

The mechanisms, which underlie the ROS-mediated MAPK activation, may involve direct alteration of kinases and transcription factors, and indirect modulation of

cysteine-rich redox-sensitive proteins such as thioredoxin and glutathione S-transferase

ASK1, an upstream regulator of MAPK, is inhibited in non-stressed cells through its association with thioredoxin (Saitoh et al., 1998) Increased ROS levels lead to dissociation of this complex and thereby enable the activation of ASK1 and downstream MAPKs (Liu et al., 2000) In the case of JNK, ROS triggers the detachment of JNK associated Glutathione-S-transferase-π (GSTp) and thereby facilitating JNK activation

(Adler et al., 1999) ROS-dependent activation of JNK may also involve the

downregulation of a JNK phosphatase (Chen et al., 2001) However, apart from these, additional mechanisms linking ROS and MAPKs may also exist

1.5.3 PUFAs and MAPKs

Studies looking into relationship of PUFAs and MAPK are not many In rat liver epithelial WB cells and Jurkat T cells, activation of MAPK by PUFA appears to be a general response, as both the n-3 and n-6 PUFAs activated MAPKs (Hii et al., 1995; Hii

et al., 1998) In contrast, in Mesangial cells the activation was selective as only DHA but not EPA activated JNK (Yusufi et al., 2003a) Among the n-6 PUFAs, arachidonic acid has also been reported to activate JNK but the activation seems to be cell specific Arachidonic acid activates JNK in stromal and rabbit proximal tubular epithelial cells and Jurkat T cells, but not in neutrophils (Hii et al., 1998) The cell specific effects of arachidonic acid are also seen in haematopoietic cells as arachidonic acid activates JNK and p38 in H7.bcr-abl A54, Jurkat and RPMI 7666 cells but not in U937 cells (Rizzo et

al., 2002) In contrast, PUFAs are also reported to have an inhibitory effect on MAPK activation, as in Jurkat T cells, DHA and EPA inhibited Phorbol Myristate Acetate (PMA)-stimulated ERK activation via Protein Kinase C-dependent and –independent pathways (Denys et al., 2002; Denys et al., 2001) This indicates that some of the actions

of PUFA could be mediated via the MAPK cascade

1.6 Peroxisome Proliferator-Activated Receptors (PPARs)

1.6.1 Functions of PPARs

The peroxisome proliferator-activated receptors (PPARs) are a subgroup of ligand-activated nuclear receptors responsible for the regulation of cellular events ranging from lipid homeostasis to cell differentiation and apoptosis (Rosen and Spiegelman, 2001) There are 3 known subtypes of PPARs – PPARα, PPAR β/δ and PPARγ PPARα

is mostly present in the liver, kidney, heart and muscle and is involved in the catabolism

of fatty acids (Braissant et al., 1996) PPAR β/δ is the most widely distributed subtype and plays a role in placentation and adiposity (Barak et al., 2002) The most widely studied form is PPARγ, which is widely expressed in a variety of cell types, including adipocytes, macrophages and colonocytes, and plays a critical role in regulating adipocyte differntiation (Willson et al., 2001)

The three PPAR isoforms (α,δ,γ) differ in their C-terminal ligand binding domains, and each appears to bind and respond to a specific subset of agents including hypolipidemic drugs, long chain fatty acids, prostaglandins, and antidiabetic thiazolidinediones Upon ligand binding, the receptor becomes activated and an active heterodimer of PPAR and the retinoid X receptor bind to DNA (Murphy and Holder,

2000) Binding of the complex to gene regulatory sites on the DNA, termed ‘Peroxisome proliferators Response Elements’ (PPRE) leads to increased transcription of numerous genes Target genes induced include those involved in growth regulatory pathways, lipid transport and storage and also in colon cell maturation and immune modulation (Willson

et al., 2001; Gupta et al., 2001)

1.6.2 PPARγ and colon cancer

Studies of humans, animals and cultured cells support the view that modulation of PPARγ activation may have therapeutic benefits (Murphy and Holder, 2000; Sporn et al., 2001; Gupta and DuBois, 2002) Apart from adipocytes, PPARγ is abundantly expressed

in the colon (Fajas et al., 1997) with the highest level of receptor expression observed in postmitotic, differentiated epithelial cells facing the lumen (Lefebvre et al., 1999) Consistent with this expression pattern, mutations of PPARγ have been reported in colon cancers (Kinzler and Vogelstein, 1996) Each of the cancers with the mutant PPARγ also had a normal PPARγ allele This suggests that even a partial loss of PPARγ function enhances the transformation to colon cancer Human colon cancer cell lines with a point mutation in the PPARγ gene fail to undergo differentiation or growth inhibition even in the presence of agonists (Gupta et al., 2003) But still, the role of PPARγ in colon cancer

is controversial, as both the suppression and enhancement of tumor growth have been observed in different animal models treated with activators of PPARγ In nude mice, treatment with PPARγ ligands resulted in a reversal of events associated with colon cancer, and yielded smaller tumors (Sarraf et al., 1998) But in contrast, in a Min+/- mice model, which lacks one functional copy of the APC tumor suppressor gene thus

predisposing them to cancer, treatment with PPARγ ligands, troglitazone and rosiglitazone- resulted in an increased number of tumors (Lefebvre et al., 1998; Saez et al., 1998) These conflicting results have been partially explained by the demonstration that in mice with mutated APC gene, PPARγ loses the ability to regulate colon tumorigenesis, whereas in wild type APC mice, PPARγ functions as a tumor suppressor gene (Girnun et al., 2002) This is relevant in human cancers because mutation of the APC gene is often the initiating event for the majority of sporadic colorectal tumors (Powell et al., 1992) Clearly, additional studies are required to determine the role of ligands of PPARγ in both the prevention and treatment of colon cancer

1.6.3 PUFAs and PPARs

Fatty acids have been shown to bind and to activate PPARs (Desvergne and Wahli, 1999) The fatty acids bind all the three PPARs, with PPARα exhibiting the greatest affinity Compared with the saturated and monounsaturated fatty acids, PUFAs generally have a higher affinity for the receptors (Kliewer et al., 1997) Among the PUFAs, DHA can preferentially activate PPARγ more than LA As for PPARα, and –δ, the activation was comparable (Yu et al., 1995) That PUFAs act as activators of PPARs has led many to view PPARγ as the link between dietary fat and colorectal cancer (Willson et al., 2001) It has been proposed that the increase in colonic polyps seen with the PPARγ ligand treatment is consistent with the increase in polyp burden observed when these mice were placed on a high-fat diet (Lefebvre et al., 1998; Saez et al., 1998) However there were significant differences between the two types of treatment PPARγ ligands induced polyps only in the large intestine compared to the high-fat diet which

increased the polyp number in both the small and large intestines, suggesting that fatty acids may activate other pathways also in addition to PPARγ activation That fatty acids can also act as hormones that control the activity of transcription factors proves that fatty acids are not merely energy providing molecules but are also metabolic regulators This finding opens novel perspectives for the in depth understanding of the therapeutic role of fatty acids

1.7 Objectives of the study

Evidence from epidemiological studies and animal models suggest that n-3 PUFAs have a protective effect on colon cancer But the mechanisms of the chemotherapeutic and chemopreventive actions of n-3 PUFAs are still unknown DHA, abundantly found in fish oil, is one the principal n-3 PUFAs The aim of this study is to explore the molecular mechanisms of the effects of n-3 fatty acids, and DHA in particular, in colon cancer cell lines The specific aims are summarized as follows:

(1) To look into the effects of saturated, monounsaturated, and n-3 and n-6 polyunsaturated fatty acids on cell proliferation and viability of colon cancer cell lines

(2) To study the apoptotic effects of the n-3 PUFA, DHA in the colon cancer cell line, HT-29

(3) To elucidate the effect of the MAPK signaling pathways in DHA-induced apoptosis

CHAPTER II MATERIALS AND METHODS

2.1 Cell lines and chemicals

Human colon cancer cell lines HT-29, HCT-116, SW480 and Colo205 were purchased from American Type Culture Collection (ATCC, Rockville, USA) All fatty acids, culture media (McCoy 5A, DMEM, RPMI1640), fatty acid free-bovine serum albumin, Triacsin C, etomoxir, ciglitazone, BADGE, and human recombinant TNF-α were from Sigma (St Louis, MO) Antibodies against PARP, caspase-3, c-jun, phospho-c-jun (Ser 63), p38 MAPK, phospho-p38 MAPK, p44/42 MAPK and phospho-p44/42 MAPK were from Cell Signaling (Beverly, MA) Antibodies against JNK1, COX-2, PPARγ and tubulin were from Santa Cruz (Santa Cruz, CA) The secondary antibodies (horseradish peroxidase conjugated goat anti-mouse and goat anti-rabbit IgG) and the enhanced chemiluminescence (ECL) substrate were from Pierce (Rockford, IL) The Caspase-3/7 Assay Kit was purchased from Promega (Madison, USA) The pan-caspase inhibitor z-VAD-fmk, JNK inhibitor SP600125; the p38 inhibitor SB203580, and the p44/42 inhibitor PD98059 were purchased form Calbiochem (San Diego, CA) 2’7’-dichlorofluorescein diacetate (DCFH-DA) was purchased from Molecular Probes, Inc (Eugene, OR, USA) pNF-κB-luc vector (Mercury Pathway Profiling System) was from Clontech (San Diego, CA) Research Biolabs synthesized PPARγ sense and antisense oligonucleotides Luciferase assay kit was from Promega (Madison, WI) Lipofectamine transfection reagent was from Invitrogen (Carlsbad, CA) Radioactive probe γ-P-32 was from Perkin-Elmer (Shelton, CT) The pDsRed expression vector was obtained from Clontech (Palo Alto, CA) The dominant negative JNK1 and JNK2 vectors (DN-JNK1, DN-JNK2) were kindly provided by Dr A.G Porter (Li et al., 2004)

2.2 Cell culture and fatty acid supplementation

HT-29 and HCT-116 cells were cultured in McCoy 5A; SW480 in DMEM; and Colo205 in RPMI 1640 medium supplemented with 10% FBS and 100 units/ml penicillin and 100 µg/ml streptomycin All cells were incubated in a humidified atmosphere of 5%

CO2 at 37o C In all experimental procedures, cells were serum starved for 12 hours before chemical treatment Fresh serum free medium was then added for various designated treatments

For fatty acid supplementation, all fatty acids were dissolved in absolute ethanol

In order to minimize the oxidation of fatty acids, stock solutions (50 mM) were stored under nitrogen at -80oC The necessary aliquot of the stock solution was combined with fatty acid free-bovine serum albumin (BSA) at a molar ratio of 1:1 in serum free medium

In all experiments, the control group was supplemented with the same concentration of fatty acid free-BSA in ethanol The fatty acid-albumin complex solution was freshly prepared and filtered before use All aliquots of fatty acids were used within a week

2.3 Determination of cell viability – MTT assay

Tetrazolium dye- 3,(4,5-dimethylthiazol-2-yl)2,5-diphenyl-tetrazolium bromide (MTT) colorimetric test was used to monitor cell growth indirectly as indicated by the conversion of the tetrazolium salt to the colored product, formazan, the concentration of which was measured spectrophotometrically (Hansen et al., 1989) Cells were seeded in 96-well microplates at a density of 1x104 cells/well in100µl medium After incubation with PUFAs in serum-free medium for the indicated time periods, 25 µl of MTT solution (5 mg/ml) was added to each well After 2 hours, 100 µl of lysis buffer (50% DMF and